Jan 28, 2022 · We propose a novel graph neural network to estimate extended persistence diagrams (EPDs) on graphs efficiently.

Extended persistent homology captures 0-dimensional (connected components) and 1-dimensional (loops) topological structures and summarizes their topological.

However, computing extended persistent homology summaries remains slow for large and dense graphs and can be a serious bottleneck for the learning pipeline.

The proposed neural network aims to simulate a specific algorithm and learns to compute extended persistence diagrams for new graphs efficiently. Experiments on ...

This work proposes a novel graph neural network to estimate extended persistence diagrams (EPDs) on graphs efficiently, built on algorithmic insights, ...

We propose a novel learning method to compute extended persistence diagrams on graphs. The proposed neural network aims to simulate a specific algorithm.

Nov 15, 2022 · Topological features based on persistent homology can capture high-order struc- tural information which can then be used to augment graph ...

Accelerating Extended Persistent Homology. In general, computing extended persistent homology relies on the well-known matrix reduction algorithm [8]. Much ...